摘要
本文提出了一个新的用于估计化学体系组分的化学计量学方法,方法的基础在于具有化学意义的特征子空间对微扰和数据的旋转变换具有较大的稳定性,通过观察多次实验中投影残差的标准方差来确定化学秩。对实际化学体系的分析结果表明,该法为组分数的估计提供了一个有价值的工具。
Based on the invariability of the subspace to the noise addition and linear transformation to the raw data, a new method is proposed for estimation of chemical rank of two-way data matrices. The rank can be determined by inspection of the standard error of projection residual in several experiments. The results of two real chemical systems show that the method provides a useful tool for estimation of chemical component number of two-way data matrice.
出处
《分析科学学报》
CAS
CSCD
2005年第2期185-187,共3页
Journal of Analytical Science
关键词
子空间
秩估计
化学计量学
Subspace
Rank estimation
Chemometrics
